[reportlab-users] Reportlab problem displaying japanese characters in a table

[Andy Robinson]

2009/6/12 Michael Dorrian <michael.dorrian at gmail.com>:

> Thanks for the replys...........

> Here is my function. I am just showing u what i had to do to display

> the first field of the table correctly from the returned rows of the

> sql.............there must be an easier way than this. The field

> headings display correctly. Its just the database data thats the

> problem.


You said earlier that you get an error message when running this.  Can
you post the traceback?   If it's in the table-building line, can you
put this on a separate line....?


str(..) requires its arguments to be ASCII, so whatever the content
encoding is in JPYRows[0][11], if it contains any non-ascii, it will
fail.   Try getting the first field directly and converting it from
utf8 to unicode on a separate line and then assembling it.  We'll then
know where it fails.  e.g.

db_content = JPYRows[0][11]
unicode_content = db_content.decode('utf8')
t = Table(Paragraphs(...headers..., (unicode_content, ....)))


...if that fails, try the other Japanese encodings in plave of 'utf8'
- do you know anything about EUC or Shift-JIS?





BTW, all those steps you are going through are necessary.  If you are
doing them in several parts of your program, you can hide them in a
function somewhere to make it more readable; every project I work on
tends to have a database connection already passed around, and I have
a utility function something like this...

def query(conn, sql_statement):
    cur = conn.cursor()
    cur.execute(sql_statement)
    rows = cur.fetchall()
    cur.close()     #good idea with some DB APIs.
    return rows

...so the actual code I'd work on from project to project looks more like...


table_data = [[headers]] + query(conn, 'SELECT blah blah....')
t = Table(table_data)
story.append(t)


-- 
Andy Robinson
CEO/Chief Architect
ReportLab Europe Ltd.
Media House, 3 Palmerston Road, Wimbledon, London SW19 1PG, UK
Tel +44-20-8545-1570